Good tip. I changed it to 512 also, yesterday.
Lemmy
Everything about Lemmy; bugs, gripes, praises, and advocacy.
For discussion about the lemmy.ml instance, go to !meta@lemmy.ml.
As a user, I just woke up and can say Lemmy.world feels waaay smoother on both desktop and mobile, interacting with other instances too for absolute sure.
my server is just me currently.... but it's got 100GB of RAM and 30CPUs so i kicked my value up to 200k but this shit (the fediverse) is still slow as hell/doesn't sync with most other servers because their specs are so low. People need to stop running them on $4 VPS shit boxes.
And that only effects outbound federation basically? So a small instance shouldn't have much issues with this, even if it's subscribed to a lot of very busy communities?
Correct
@nutomic I wonder if my mastodon instance requires similar tweaking, I haven't seen any new updates from Lemmy.ml today on my feed...
I have a bunch of lemmy.ml communities still stuck in "Subscribe Pending" state even after waiting for days. Canceling and re-subscribing does not help.
Ah maybe that should be mentioned somewhere that this specific config requires a restart of the Lemmy backend.
I think settings that do require this should be only in the lemmy.hjson config file.
I increased ours to 128 yesterday, but wanted to keep downtime low so I didn't restart.
Going from 512 to 160,000 is a massive parameter change.
Network replication like this presents a ton of issues with servers going up and down, database insert locking the tables, desire for backfill and integrity checks, etc.
Today things are going poorly, this posting has an example: https://lemmy.ml/post/1239920 -- comments are not showing up on other instances after hours.
From a denial-of-service perspective, intentional or accidental, I think we need to start discussing the protocol for federation. When servers connect to each others, how frequently, how they behave when getting errors from a remote host.
Is lemmy_server doing all the federation replication in-code? I would propose moving it to a independent service - perhaps a shell application - and start looking at replication (and associated queues) as a core thing to manage, optimize, and operate. It isn't working seamlessly and having hundreds of servers creates a huge amount of complexity.
Thanks for the heads up.
I did notice that syncing has been slow lately, hopefully this does fix it.